Don Nolan-Proxmire Headquarters, Washington, DC Nov.

12, 1996 (Phone: 202/358-1983) Keith Henry Langley Research Center, Hampton, VA (Phone: 804/864-6120) RELEASE: 96-233 AIRCRAFT FLIGHT CONTROL SYSTEM WINS "BEST OF WHAT'S NEW" An experimental aircraft flight control system that learns as it flies has been honored as one of the best technology developments of 1996. Developed for NASA and the U.S. Air Force, the computerized flight control system has been installed on an 8-foot-4-inch remotely piloted aircraft called"LoFLYTE" being prepared for flight demonstrations this month. The jetpowered aircraft was developed by Accurate Automation Corp., Chattanooga, TN, under the Small Business Innovation Research program. The LoFLYTE hypersonic waverider aircraft was named one of the 100 "Best of What's New" in the annual Popular Science magazine competition. Winners were announced today at an exhibition in New York City's Central Park and will be featured in the magazine's December issue. The experimental LoFLYTE aircraft will be used to explore new flight control techniques involving neural networks, which allow the aircraft control system to learn by mimicking the remotelysited pilot. Technologies being implemented in the LoFLYTE program could eventually find their way into commercial, general aviation and military aircraft. Flights are taking place at Edwards,CA, with the support from NASA¹s Dryden Flight Research Center. The model is a Mach 5 waverider concept, a futuristic

hypersonic aircraft configuration that could cruise on top of its own shockwave if powered to reach hypersonic speeds. LoFLYTE represents the first known flying waverider vehicle configuration. In the current flight tests it is powered by a small-scale jet engine and will reach subsonic speeds to explore takeoff and landing control issues. The aircraft has been designed to demonstrate that neural network flight controls are superior to conventional flight controls. Neural networks are computer systems that actually learn by doing. The computer network consists of many interconnected control systems, or nodes, similar to neurons in the brain. Each node assigns a value to the input from each of its counterparts. As these values are changed, the network can adjust the way it responds. The LoFLYTE aircraft's flight controller consists of a network of multiple-instruction, multiple-data neural chips. The network will be able to continually alter the aircraft's control laws in order to optimize flight performance and take the pilot's responses into consideration. Over time, the neural network system could be trained to control the aircraft. The use of neural networks in flight could help pilots of future aircraft to fly in quick-decision situations and help damaged aircraft land safely even when controls are partially disabled. The waverider was chosen as the testbed for the neural networks because the configuration has an inherently high lift-to-drag ratio at hypersonic speeds. If neural networks can control this "worst-case scenario" configuration, then they should be able to handle virtually any conventional configuration, project officials say. Individuals and organizations cited for their work with LoFLYTE are James L. Hunt, NASA Langley Research Center, Hampton, VA; Dr. Kervyn Mach, Air Force Wright Laboratory Dayton, OH; and Bob Pap, Accurate Automation. - end -